LC-MS-based proteomics often relies on data-dependent acquisition (DDA) for quality control. Here, the authors demonstrate that data-independent acquisition (DIA) outperforms DDA in detecting subtle changes in LC-MS status in large-scale quantitative proteomics experiments. They further prioritized 15 QC metrics and developed an AI model, implemented in a free software called iDIA-QC, for detecting LC-MS faults.
- Huanhuan Gao
- Yi Zhu
- Tiannan Guo